J 094 



Z=250 y oh/cm, so that ij,nnx was equal to ^^^. Maximally an 



1 



750 

 electron loses therefore by the impact with an argon atom the 



part of the energy it has gained between two succeeding iin- 



750 



pacts. In this case the loss of energy may be entirely neglected and 



we may assume that the energy of an electron it determined only 



by the way it has travelled in the direction of (he (ield viz. ail 



electrons at the same distance from the cathode will have the same 



velocity. 



In measurements with constant plate distance this is not at all 

 the case. As long as the potential difference between the plates is 

 small, the field is weak and y therefore great. In this case we 

 must therefore always take into account the energy loss by the 

 impacts and especially for the first steps of the current tension-curve 

 this influence will be considerable. For this reason measurements 

 with a constant practically chosen field are preferable for the deter- 

 mination of the ionisation potential. 



As to the height of the steps, according to Kranck and Hertz 

 they must be in the ratio 1:2:4 etc. When we disregard the 

 round corners the current could be represented as a function of the 

 tension by a formula of the form / =: «^ A', 2", where .V^ is written 

 for the number of the electrons emilted by the cathode and n for 



V 



the greatest integer smaller than — . {V is the potential difference 



between the two plates). 



In our measurements the ratio 2:1 never occurred, but 1.3:1 

 and 1.5:1. Doubtlessly this is due to the influence of impurities in 

 the argon. By these the deviations are also greater in the measure- 

 ments with a constant field than in those with a constant plate 

 distance. It seems therefore to be of importance to repeat these mea- 

 surements still with a purer gas. Finally we can conclude from the 

 curves with a constant field, that nearly every impact of an electron 

 that has travelled through 12 Volt causes ionisation. From the 

 preceding calculation it is found, that the mean value of the path 

 travelled by an electron with a velocity corresponding to 12 Volt 

 between two impacts in the direction of the field is 0,001 cm. The 

 measurements teaches however that all ionisations take place in a 

 layer of maximally 0,004 cm., so that nearly each impact that can 

 cause ionisation reallv does cause it. 



